International Journal of Advanced Mechatronic Systems (15 papers in press)
Research on Static Reactive Power Generator Based on Asymmetric Distribution Network
by Fuzhuan Wu, Shengjun Wen, Sheng Peng
Abstract: Negative or zero sequence components are generated when the voltage is asymmetric or harmonic in distribution networks. Meanwhile, the decoupling process of traditional dq transform is complex. To solve the above problems, firstly, the T/4 delay method (T is the period of grid voltage) is presented to separate positive and negative sequence components, which improves the stability of software phase-locked. Then, Software Phase-Locked Loop (SPLL) is designed to ensure the instantaneity of reactive current check. Besides, a double-loop control scheme combining proportional integral (PI) controller for DC voltage outer loop and proportional resonance (PR) controller for AC current inner loop without decoupling is designed by considering the characteristics of traditional PI and PR without static error regulation. It avoids the complicated decoupling process and improves the real-time performance of the system. Finally, both simulation and experimental results are given to verify the feasibility of design scheme in the Static Var Generator (SVG) system by MATLAB/Simulink and experimental platform based on DSP28335.
Keywords: Static reactive power generator; asymmetrical distribution network; PR controller; software phase-locked loop.
Data-based Reinforcement Learning for Lane Keeping with Input Saturation
by Rui Luo, Dianwei Qian, Qichao Zhang
Abstract: With the development of artificial intelligence, autonomous driving has received extensive attention. As a very complex integrated system, the autonomous vehicle has several modules. This paper is related to the control module, which is used to design an optimal or near-optimal controller to control the desired trajectory of the vehicle. In this paper, lateral control strategy for lane keeping task is proposed based on the model-free reinforcement learning. Different from the model-based methods such as linear quadratic regulator and model predictive control, our method only requires the generated data rather than the perfect knowledge of the system model to guarantee the optimal performance. At the same time, in order to meet two needs of passengers' comfort and fuel economy, input saturation should be considered in the design of the control module. A low-gain state feedback control method is adopted. It mainly solves some algebraic Riccati equations for data-based lateral control. Finally, the corresponding simulation is given and the validity of the algorithm is verified.
Keywords: lateral control; lane keeping; input saturation; reinforcement learning.
A Self-Learning Fall Detection System for Elderly Persons Using Depth Camera
by Xiangbo Kong
Abstract: The machine learning revolution is redesigning modern health care, and with the growth of the elderly population, fall detection has become an important research topic in health care. This paper surveys advances in machine learning-based fall detection technologies and reviews sensor-based, image processing-based, and wearable sensor-based fall detection systems and applications. In addition, this paper proposes a self-learning posture analysis and eye status-based fall detection system to solve the issue of mis-detections in fall detection systems, which have not been addressed in past works. Furthermore, this work proposes an image-feature-separation system that can use image processing with a low risk of privacy disclosure. Moreover, this work establishes a data set, which includes 36 non-fall/fall cases comprising 25,200 images that can be used not only for this research but also in related studies. Experimental results show that this system can detect a fall with high accuracy and solve mis-detections in machine learning-based fall detection systems.
Keywords: Health care; elderly persons; fall detection; self-learning; posture analysis; eye status; support vector machine.
A Learning-based Short-term Wind Speed Forecasting Approach through Spiking Neural Networks
by Jing Hu, Lili Xie, Yixin Chen, Weidong Liu, Xingpeng Zhang, Dianwei Qian
Abstract: In real-world applications, the index of wind speed is concerned to many fields, such as power generation, flight vehicles and even weather forecasting. Especially, the index plays an extremely important role in wind power systems. Not only is the index related to the economy, but also it has direct effects on the security. Unfortunately, wind is originated from the weird nature and its speed is inherently stochastic. It is already hard enough to accurately measure the wind speed. Its forecasting undoubtedly becomes harder and more challenging. The long-term wind speed forecasting is confronted with the dilemma of computational burden and forecasting accuracy so that a substantial part of wind speed forecasting is short-term in reality. This paper focuses on the problem of short-term wind speed forecasting. Empirically, wind speed is relevant to temperature, pressure, humidity and some other factors. It is too complex to model the wind speed by mathematical formulas. The technique of neural networks is a learning-based approach. By this technique, the method of spiking neural networks is one of the most successful methods to fulfill the modeling of complex dynamics and the exploitation of learning ability. With the purpose of refining the forecasting accuracy and expediting the learning speed, this paper investigates a spiking-neural-network-based structure, designs a hybrid learning algorithm that combines the adaptive learning rate and the momentum term and implements them for the short-term wind speed forecasting. Experiments and comparisons are illustrated to show the effectiveness and feasibility of this learning-based forecasting approach.
Keywords: wind speed; forecasting; short term; spiking neural networks; learning algorithm; modelling; power generation.
Fingerprint and password controlled garage access system with belt pulley and power screw driven mechanism
by Md Mostafizur Rahman Komol, Md Karimul Joarder, Abdullah Arafat, Amit Kumer Podder
Abstract: Garage access automation and security have been a favored concern with technological advancement. In this research paper, garage access security and automation are maintained with the affiliated control of the biometric fingerprint recognition system and password verification system. Fingerprint recognition is only for the restricted purpose used by registered vehicle owners of the garage. Everyone is permitted to access through a password verification system. The affiliation of the dual-sensing system is intended to maintain robust security and retaining the record of any drivers access without the vehicle owners. This detection is presented in a liquid crystal display. Moreover, the system is enriched with a subtle design of the access door and its operation mechanism. Screw-driven door operation is utilized to convert the rotating motion of the motor to linear motion and slide the door open or close. The requisite motor torque requirement for door control by the screw operation is calculated and verified. Also, belt pulley is adjoined with the screw and rotor for better power transmission efficiency. Furthermore, for the reduction of the friction by the door, roller attachment, as well as subtle celerity of garage door, is aptly maintained during operation.
Keywords: Fingerprint; Password; Garage access; Belt-pulley; Power-screw; Motor-torque.
Operator-based nonlinear fault detection and fault tolerant control for microreactor using one-class SVM
by Yoshiki Ogihara, Mingcong Deng
Abstract: This paper proposes a method of nonlinear fault detection scheme using one-class Support Vector Machine (SVM) and operator-based fault tolerant control system for a microreactor with Peltier device. This method requires only one-shot training data for the normal condition of the plant. The data is used to learn for the SVM, it realizes nonlinear fault detection without the known fault condition of the plant. Also, the control system that consists of the nonlinear model of the microreactor system and operator-based controller designed by the model is introduced as the fault tolerant control system using compensation operator to remove the sensor fault. The usefulness of the method is shown by an experiment of the temperature sensor fault case for the temperature control system of the microreactor system.
Keywords: nonlinear control; fault detection; fault tolerant control; one-class classification; Support Vector Machine.
Special Issue on: RANE 2019 Intelligent Mechatronic Systems and Additive Manufacturing
Obstacle Avoidance System and Wireless Communication for an Unmanned Underwater Vehicle for Low Depth Water Surfaces
by Arockia Selvakumar Arockia Doss, Vivek Ghodeswar
Abstract: An underwater glider is a most commonly used unmanned underwater vehicle but it has limitations to avoid obstructions in its path. To overcome this problem, the application of obstacle avoiding system is needed. This paper describes the development of an unmanned underwater vehicle (UUV) with integration of sensor-actuator network to avoid obstacles. To study the hydrodynamic behavior of the proposed UUV, computational fluid dynamics (CFD) is carried out by considering pure surge and heave motion. The UUV is equipped with obstacle avoidance system with Infra-Red (IR) sensor and wireless communication module. Experimental tests are conducted to understand the behavior of the UUV in low depth water surfaces and also to validate the CFD simulation results. The UUVs development, motion analyses and preliminary tests in obstacle avoidance are reported.
Keywords: underwater vehicle; CFD; motion study; thruster; IR sensor; sensor-actuator network.
Maze Path Planning of Mobile Robots by Gradient Map Rendering and Gradient Follow
by Arockia Selvakumar Arockia Doss, Arka Das, Pavan K L, Dinakaran D
Abstract: Path Planning for a human being is very easy to reach a desired location in a room, by avoiding obstacles on the way, by generating a mental map and uses this map to find the optimal path. This is a difficult task in case of a robot. In order to make the robot adapt, the system is fed with different obstacle arrangement in the same room and allow the robot to finalize the optimal path avoiding the obstacles. To achieve, the gradient map rendering algorithm is proposed with successful simulation results in MATLAB. The new map produced has given a gradient again using the same algorithm. After the rendering process is completed, the robot climbs up or down the gradient using the maximum or minimum local gradient technique respectively, to find its way to the destination cell. Gradient surface plots are obtained for a variety of mazes to give a visualization of how the gradient is being formed. Results are obtained after maze simulation successfully shows the most optimized path in any kind of maze.
Keywords: gradient follow; gradient map rendering; gradient movement; grid maze; MATLAB; mobile robot; path planning.
Structural Design and Analysis of a Lower Limb Exoskeleton for Elderly
by Vishnu Vardhan Dadi, Sathwik P. V. N. S, Mahesh D, Jaswanth Dala, Karthik Kumar S, Ramya M. M, Dinakaran D
Abstract: Rehabilitation of the elderly is often limited to restoration of the ability to perform as many activities of daily living. Mobility is identified as most essential rehabilitation required for elderly. Sit-to-Stand (STS) manoeuvre is a common aspect of mobility. In this paper, a lower limb exoskeleton is designed to assist elderly during STS cycle. The design of lower limb exoskeleton is tested for its structural strength. The mechanical design of the exoskeleton can adapt to varying body shapes (height, weight and waist circumference) of elderly. Static structural analysis for stand position is carried out in Ansys Workbench to find whether the design can withstand a maximum load during the static condition. Modal Analysis was done to find the natural frequency vibration of the design and the deformation of the exoskeleton with respect to the mode vibrations.
Keywords: Rehabilitation; Exoskeleton; Lower Limb; Elderly; Structural Analysis; Modal Analysis; Gait Analysis.
Investigation on Dual Nozzle Fused Deposition Modelling using Industrial Robot
by Sri Harsha Arigela, Ch R. Vikram Kumar
Abstract: Fused Deposition Modeling (FDM) process builds the 3D objects by extruding the material in layer-by-layer fashion from bottom to top. In conventional FDM machine due to limited degrees of freedom, change in extruder orientation is not possible. Industrial robots with six degrees of freedom are capable to move and orient the end effector in 3D space with high resolution and repeatability. The limitation of conventional FDM machine can overcome by using an industrial robot with a dual nozzle extruder as an end effector. In this work, an attempt is made to print objects with an industrial robot with dual nozzle extruder attached to it. The results showed that the orientation of the extruder can be changed easily, so that a strong bonding between the layers is possible while going for dual nozzle extrusion. By extruding the same material with both nozzles, the printing time is less than the conventional FDM process. The methodology adopted for printing the objects using ABB Robotstudio
Keywords: Industrial Robot; Fused Deposition Modeling; Toolpath Generation;.
Platform Tilt Stabilization Using Inertial Measurement Unit Sensor
by Prasad Elumalai, Thirumal Azhagan Murugan, Karthikeyan Palanisamy
Abstract: This paper aims to develop a low cost two-axis pan-tilt platform stabilization setup using a low-cost IMU (Inertial Measurement Unit) sensor, utilize elegant and widely used sensor fusion and control algorithm, and demonstrate the performance of such a system. To perform platform stabilization, reliable pan-tilt angular estimates are required from the IMU sensor. For this purpose, techniques such as sensor bias removal, Digital Low Pass Filtering (DLPF) and sensor fusion algorithms are deployed. FIR (First-order Impulse Response) filter is chosen as the DLPF algorithm and the Complementary Filter (CF) is chosen as the sensor fusion algorithm as these are widely recognized for their need of very less computational power. Finally, the pan-tilt stabilization is performed by two separate PID servo control tuned using Zeigler-Nichols manual tuning rules and finally the performance of the control system in tracking the angular estimates is finally studied.
Keywords: Pan-Tilt Platform; Platform Stabilization; Inertial Measurement Unit; Sensor Fusion; Complementary Filter; PID Servo Control.
Design and Development of Robotic End-effector Position Measuring Device
by Aashith C, Muralidhara Rao
Abstract: Robot end-effector position measurement is very essential to determine the error between the commanded position and actual position reached by the robot end-effector. This also ensures the error with which the robot end-effector traces the commanded path. Conventionally position measuring devices like laser trackers, Co-ordinate Measuring Machine and other non-contact based position measuring devices are widely used to measure the position of the end-effector due to their high accuracy and precision, even though they are very expensive. This paper presents a cost effective solution for the measurement of position of robot end-effector. An end-effector position measuring device is designed and developed in spherical coordinate system using two absolute rotary encoders and a draw-wire sensor. A mathematical model for the end-effector position measuring device is developed to determine the position of the end-effector with respect to a reference coordinate system using 3-D homogeneous transformation approach. End-effector position measuring device is tested for ABB IRB 1600 for numerous poses and for a straight line path. The newly developed end-effector position measuring device is found to be capable of measuring the end effector position with an accuracy of 2 mm and hence can be implemented in calibration of industrial robots.
Keywords: end-effector position measuring device; 3D Homogeneous transformation; kinematic model; end-effector position/path error.
Machine Learning based Ovarian Detection in Ultrasound Images
by Kiruthika V, Sathiya S, Ramya M.M
Abstract: Computer aided ovarian detection and ovarian classification is important in infertility treatment in women. In the proposed methodology, an intelligent automatic detection and ovarian classification with grading based on integration of intensity and texture features using artificial neural network is developed. Three texture features such as autocorrelation, sum average and sum variance obtained from gray level co-occurrence matrix (GLCM) and intensity obtained using k-means clustering were fed as input to the multilayer feedforward backpropagation network for ovarian detection. Ovarian morphology was used for classification and grading of ovary. This novel technique helps the physician to grade the follicle/cyst. Performance metrics like Sensitivity, Specificity, Accuracy, Precision, F-measure, Mathews correlation coefficient and Receiver Operating Characteristic Curve were used to prove the effectiveness of the proposed Machine learning based Ovarian Detection (MLOD). The MLOD classifier yielded an average detection accuracy of 96% which is an increase of 2% as compared to the combined texture and intensity based ovarian classification (TIOC) algorithm.
Keywords: Colour space transform; Discrete wavelet transform; K-means clustering; texture features; intensity based segmentation; machine learning; intelligent classifier; Artificial neural network; ovarian detection; ovarian classification.
Development of Pass - Through Augmented Reality Interface for Human Robot Interaction
by Madhumitha G, Nandhini M, Senthilnathan R
Abstract: In this technology driven world, human machine interaction is indispensable. Conventional Human Machine Interface techniques pose the challenges of extensive training and increased cognitive load. Usage of hand gestures, face and speech recognition, eye ball tracking, etc are in the upward trend to overcome these limitations and prove to be more intuitive and user friendly. In this paper, a vision based pass-through augmented reality system is developed that enables the user to command the mobile robot for various mobility and navigation tasks. A Pass-through augmented reality setup is achieved by a combination of a typical Virtual Reality headset and a stereo camera with depth perception. The stereo camera and a Leap Motion sensor mounted on a wearable VR headset acts as the vision system for the user and the gesture recognition system respectively. Necessary graphical user interface with the list of functions and information will appear on the VR headset display from which the user can select the option/command to be given to the mobile robot. The gesture made by the user in the GUI is detected and communicated to the robot through wireless means to perform the corresponding task.
Keywords: Mixed Reality; Augmented Reality; Human-Machine Interaction.
Special Issue on: RANE 2019 Intelligent Mechatronic Systems and Additive Manufacturing
Design and tuning of Fractional order Model based control for higher order process using Bat algorithm
by Hemavathy P R, Mohamed Shuaib Y, Lakshmanaprabu S K
Abstract: Fractional order system gained more popularity in the engineering domain due to its accuracy of representing physical system. For solving many engineering, nonlinear, multimodal problems Bat optimization algorithm is effective to obtain global optimum solutions through rapid convergence. In this paper, time domain based approximation method is applied to approximate higher order transfer function to fractional order first order plus dead time transfer function using bat algorithm (BA). The main contribution of the proposed work is that the IMC based Proportional Integral Derivative (IMC-PID) controller tuning rules are developed analytically for the approximated fractional order model and integer order model. The developed IMC-PID resulted with single tuning parameter which is tuned using BA for minimization of Integral Time Absolute Error (ITAE). The simulation comparison results demonstrated that the proposed controller design procedure is capable of controlling higher order process effectively.
Keywords: Fractional modeling; Bat Algorithm; Internal model control; Fractional PID with filter; Integer PID with filter.